Skip to content

Commit 7eaa5ad

Browse files
Updating chunking contract as per UDS recommendation
1 parent 95f9312 commit 7eaa5ad

1 file changed

Lines changed: 14 additions & 14 deletions

File tree

src/datacustomcode/function/feature_types/chunking.py

Lines changed: 14 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -31,8 +31,8 @@
3131
)
3232

3333

34-
class DocumentType(str, Enum):
35-
"""Document type enumeration"""
34+
class ElementType(str, Enum):
35+
"""Element type enumeration"""
3636

3737
TEXT = "text"
3838
TITLE = "title"
@@ -74,32 +74,32 @@ class SearchIndexChunkingV1TranscriptField(BaseModel):
7474
description="Speaker name for audio/video transcripts",
7575
examples=["Agent"],
7676
)
77-
start_timestamp: Optional[str] = Field(
77+
start_timestamp: Optional[float] = Field(
7878
default=None,
79-
description="Start timestamp in ISO8601 format: YYYY-MM-DDTHH:MM:SS.ffffff",
80-
examples=["2026-03-25T02:01:24.918000"],
79+
description="Start timestamp of the audio/video clip",
80+
examples=["1.0"],
8181
)
82-
end_timestamp: Optional[str] = Field(
82+
end_timestamp: Optional[float] = Field(
8383
default=None,
84-
description="End timestamp in ISO8601 format: YYYY-MM-DDTHH:MM:SS.ffffff",
85-
examples=["2026-03-25T02:01:30.500000"],
84+
description="End timestamp of the audio/video clip",
85+
examples=["8.75"],
8686
)
8787
model_config = ConfigDict(extra="ignore")
8888

8989

9090
class SearchIndexChunkingV1Metadata(BaseModel):
9191
"""Metadata for input documents."""
9292

93-
type: Optional[DocumentType] = Field(
94-
default=DocumentType.TEXT,
93+
type: Optional[ElementType] = Field(
94+
default=ElementType.TEXT,
9595
description=(
96-
"Document type of the chunk input. Currently only 'text' is supported."
96+
"Element type of the chunk input. Currently only 'text' is supported."
9797
),
9898
examples=["text"],
9999
)
100100
page_number: Optional[int] = Field(
101101
default=None,
102-
description=("Page number in the source document (0-based). "),
102+
description=("Page number in the source document."),
103103
examples=[1],
104104
)
105105
transcript_fields: Optional[SearchIndexChunkingV1TranscriptField] = Field(
@@ -111,7 +111,7 @@ class SearchIndexChunkingV1Metadata(BaseModel):
111111
)
112112
text_as_html: Optional[str] = Field(
113113
default=None,
114-
description=("HTML representation of the chunk text, if available. "),
114+
description=("Table represented as HTML"),
115115
examples=["<p>Online Remittance Instructions</p>"],
116116
)
117117
source_dmo_fields: Optional[Dict[str, Union[str, int, float]]] = Field(
@@ -194,7 +194,7 @@ class SearchIndexChunkingV1Output(BaseModel):
194194
examples=["Online Remittance Instructions"],
195195
)
196196
seq_no: int = Field(
197-
default=0,
197+
default=1,
198198
description=(
199199
"Sequential order of this chunk within the output "
200200
"Represents chunk ordering within the source document (1-based)."

0 commit comments

Comments
 (0)